Growing isn’t the same as being able to keep growing

For a while, almost any organization can absorb more activity by leaning on what it already has. People adapt, teams put in extra effort, and those who have been around longest make up with experience for whatever hasn’t been solved another way. It works well, it works for quite a long time, and it leaves no trace anywhere.

The invisible tax

What comes next rarely arrives with a clear signal. The company keeps selling, it may even keep growing, and from the outside everything looks reasonable. Inside, each new layer of growth starts costing more than the one before it. There is usually no single big problem to explain it: there is friction, and friction accumulates until it becomes a kind of invisible tax on growth.

Because a company’s ceiling is not always in the market. There can be demand, product and opportunity to spare, and still not enough internal capacity to absorb more without the way the company works starting to degrade. When that happens the organization keeps moving forward, but it needs more and more energy to cover the same distance. And what until then was an advantage —the closeness, the ability to improvise— starts showing its other face.

What a late decision costs

There is one concrete consequence worth looking at squarely, because that is where this gets paid for. Decisions do not get worse: they get later. And a decision that arrives late ends up in the P&L disguised as something else, almost always as a lost sale. So the useful question is not how much revenue has grown, but how much the cost of coordination has grown per euro of revenue, and how much management attention a customer incident consumes today compared with three years ago. If those curves climb alongside sales, we are buying size with future margin.

Why sending a decision up always pays off

It is worth saying something uncomfortable here, because it is what explains why this does not correct itself. Whoever ends up concentrating the decisions is not doing it because of some flaw of theirs. They are doing it because they are good, and because the organization has learned that perfectly well. For the person sending a decision up, sending it up is the best option available: they get a better decision and they stop being the one responsible for having made it. The delay, what is more, is not paid by them. Escalating costs nothing to whoever escalates; it costs one single person. That is why asking the organization to stop sending things up never works: there is nobody down there with a reason to stop.

Professionalizing is not adding bureaucracy

The usual answer is to professionalize, and that is where another misunderstanding begins. Professionalizing gets associated with filling the company with processes, probably because we have all seen organizations where the process stopped serving the business and the business ended up working to feed the process. But the opposite extreme does not work either. There are companies that look remarkably agile where that agility consists of certain people knowing how to solve, again and again, problems the system keeps producing. That is not efficiency. It is dependency, very well disguised.

The balance point is more concrete than it sounds: process should follow repeatability, never anticipate it. Installed on something that does not yet repeat, it freezes a way of working before anyone knows whether it was the right one, and that is bureaucracy. Installed on something that already repeats, it turns knowledge that lived in one person’s head into something transferable. The same decision, taken a year earlier or a year later, produces opposite results.

Something similar happens with delegation. In many companies the organization looks delegated until an important decision comes up, and then the decision goes back up. Not always out of a wish to control: often because the judgment behind those decisions has never left the head of the person who has always made them. Tasks have been delegated, but not context. And without context, decisions cannot be delegated. That is why adding management structure does not solve it on its own: if the new people still have to go up to work out what counts as a priority, we have added execution capacity and not decision-making capacity.

The 30 % test

There is one question I find more useful than any formal diagnosis for locating a company’s limit. What would happen if tomorrow it had to handle thirty per cent more activity with the structure it has today? This is not about whether the team could make an extraordinary effort for a few weeks, because it probably could. The question is what would break first. The answer forces you to point at a specific place, and that place is almost never the one you would find by looking at the org chart.

Because what separates a company that can keep growing from one that cannot is simpler than it seems. In the first, every significant dependency is a decision: somebody knows what it is, why it is kept, and what it would cost to undo. In the second they piled up on their own while things were going well, and nobody wrote it down. A deliberate dependency is an asset. An accumulated dependency is a liability recorded nowhere.

And that is the difference between growing and being in a position to keep growing. You do not read it in next year’s number. You read it in what it will take to sustain it.